Oab Bench
Saturated
Benchmark made to evaluate llms in the Brazilian Bar Examination, using a multi-judge system.
- Domain
- rl-env
- License
- unknown
- Published
- Sep 2025
Cite
Notes
Only stored in your browser.
Top score 96.5% by GPT-4.1 Mini - 1 model reporting (1 frontier)
Top models
1
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is Oab Bench?
- Benchmark made to evaluate llms in the Brazilian Bar Examination, using a multi-judge system.
- What is the current top score on Oab Bench?
- The top reported score is 96.5% by GPT-4.1 Mini, across 1 model reporting (1 from frontier labs).
- How can a model improve its Oab Bench score?
- Tools linked to Oab Bench on Sophon include OAB Bench RL Env (Kunumi) - RL environments, datasets, and scaffolds that target this eval.
- What license is Oab Bench under?
- Oab Bench is available under unknown.